Create app.py

app.py

@@ -0,0 +1,112 @@
+# Step 1: Install and import
+!pip install gradio --quiet
+
+import pandas as pd
+import numpy as np
+import string
+import re
+import gradio as gr
+
+import matplotlib.pyplot as plt
+import seaborn as sns
+
+from nltk.stem import PorterStemmer
+from nltk.corpus import stopwords
+import nltk
+nltk.download('stopwords')
+
+from sklearn.model_selection import train_test_split
+from sklearn.feature_extraction.text import TfidfVectorizer
+
+from sklearn.linear_model import LogisticRegression
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.svm import LinearSVC
+
+from sklearn.metrics import accuracy_score, confusion_matrix
+
+# Step 2: Load data
+true = pd.read_csv('True.csv', on_bad_lines='skip')
+fake = pd.read_csv('Fake.csv', on_bad_lines='skip')
+
+true['label'] = 1  # real
+fake['label'] = 0  # fake
+
+df = pd.concat([true, fake]).sample(frac=1).reset_index(drop=True)
+df = df[['title', 'text', 'label']]
+
+# Combine title and text
+df['content'] = df['title'] + " " + df['text']
+
+# Step 3: NLP Cleaning
+stop_words = set(stopwords.words('english'))
+stemmer = PorterStemmer()
+
+def clean_text(text):
+    text = text.lower()
+    text = re.sub(r'\[.*?\]', '', text)  # remove brackets
+    text = re.sub(r'https?://\S+|www\.\S+', '', text)  # remove links
+    text = re.sub(r'<.*?>+', '', text)  # remove html tags
+    text = re.sub(r'[%s]' % re.escape(string.punctuation), '', text)  # remove punctuation
+    text = re.sub(r'\n', '', text)  # remove newlines
+    text = re.sub(r'\w*\d\w*', '', text)  # remove words with digits
+    words = text.split()
+    words = [stemmer.stem(word) for word in words if word not in stop_words]
+    return ' '.join(words)
+
+df['cleaned'] = df['content'].apply(clean_text)
+
+# Step 4: Train-Test Split
+X = df['cleaned']
+y = df['label']
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+
+# Step 5: TF-IDF Vectorizer
+vectorizer = TfidfVectorizer(max_df=0.7)
+X_train_tfidf = vectorizer.fit_transform(X_train)
+X_test_tfidf = vectorizer.transform(X_test)
+
+# Step 6: Models
+models = {
+    "Logistic Regression": LogisticRegression(),
+    "Random Forest": RandomForestClassifier(n_estimators=100),
+    "SVM": LinearSVC()
+}
+
+# Train and evaluate
+results = {}
+for name, model in models.items():
+    model.fit(X_train_tfidf, y_train)
+    preds = model.predict(X_test_tfidf)
+    acc = accuracy_score(y_test, preds)
+    results[name] = {"model": model, "accuracy": acc}
+    print(f"{name} Accuracy: {acc:.4f}")
+
+# Plot confusion matrix for best model
+best_model_name = max(results, key=lambda x: results[x]['accuracy'])
+best_model = results[best_model_name]['model']
+y_pred = best_model.predict(X_test_tfidf)
+cm = confusion_matrix(y_test, y_pred)
+
+plt.figure(figsize=(5, 4))
+sns.heatmap(cm, annot=True, fmt='d', cmap='Blues')
+plt.title(f"Confusion Matrix - {best_model_name}")
+plt.xlabel('Predicted')
+plt.ylabel('Actual')
+plt.show()
+
+# Step 7: Gradio Web App
+def predict_news(text):
+    cleaned_text = clean_text(text)
+    vectorized = vectorizer.transform([cleaned_text])
+    prediction = best_model.predict(vectorized)[0]
+    return "Real News 🟢" if prediction == 1 else "Fake News 🔴"
+
+iface = gr.Interface(
+    fn=predict_news,
+    inputs="text",
+    outputs="text",
+    title="📰 Fake News Detector",
+    description="Enter a news headline or content to check if it's real or fake."
+)
+
+iface.launch()